The present invention relates generally to automatic performance devices such as sequencers having an automatic accompaniment function, and more particularly to automatic performance devices which have a function to execute note conversion based on a designated chord during an automatic performance.
Of automatic performance devices having been hitherto proposed, there is known a type which, for any of rhythm, bass and chord parts, executes an automatic performance on the basis of accompaniment pattern data stored separately from sequential performance data. In some automatic performance devices of this type, pattern numbers designating accompaniment pattern data to be used for automatic accompaniment are preset in the header of sequential performance data or by operating predetermined setting switches. Other automatic performance devices of the type contain sequential accompaniment data storing such pattern-data-designating pattern numbers in accordance with the progression of each music piece. Typically, for the bass and chord parts, such automatic performance devices are designed to conduct a note conversion process to convert the original notes on the basis of a designated chord into other notes suitable for the designated chord. The chord is designated manually by a user or operating person (operator) or by chord progression data separately stored in accordance with the progression of a music piece.
For example, U.S. Pat. No. 5,220,122 discloses that accompaniment patterns are prestored in a buffer and notes in any of the stored accompaniment patterns are converted in accordance with a note conversion table corresponding to a performed chord. This U.S. Patent fails to show that a common note conversion table is used for all the accompaniment patterns.
In the above-mentioned known automatic performance devices with such a note-converting function, however, it is preferable that a different note conversion process is carried out for each of the performance styles (e.g., rhythm styles such as pops, rock-and-roll, jazz and waltz), sections (e.g., main, fill-in, intro and ending) and parts. For example, it is preferable to apply a different pitch-limiting range of converted notes or a different note conversion table for each of the performance styles, sections and parts. In a simple form, note conversion will be executed on the basis of note-conversion-related information that is stored, along with the automatic accompaniment data, for each of the performance styles, sections and parts. Such a simple approach is however disadvantageous in that it requires a large memory capacity for storing the note-conversion-related information because such information has to be stored for each performance section and part of every performance style even where the information partly overlaps and hence can be shared among the performance styles, sections or parts.